27 Journal of Geography and Spatial Justice. Year 1th - Vol. 1 - Issue. 2 - Spring 2018

University of Mohaghegh Ardabili

Journal of Geography and Spatial Justice

Received:2018/04/10 accepted:2018/06/5

Efficiency measurements on service centers using data envelopment analysis model: A case study of counties in West Province

Issa Ebrahimzadeh*1, Diman Kashefidoost 2, Jamil Ghadermarzi3 1- Professor, Department of Geography and Urban Planning, University of Sistan and Baluchestan, , . 2- Ph.D. student, Department of Geography and Urban Planning, University of Sistan and Baluchestan, Zahedan, Iran. 3- Ph.D. student, Department of Geography and Urban Planning, University of , Yazd, Iran.

* Corresponding Author, Email: [email protected]

A B S T R A C T

Establishing service centers in counties is among the main elements in improving social welfare. Nowadays, the majority of counties not only face a shortage in service centers but also an unbalanced distribution of services. In turn, efficiency measurement of counties regarding establishing service centers is in great importance. Hence, it is essential to improve the efficiency of urban services, while distributing services in a way by which all counties are provided with balanced services. This study aims to evaluate the efficient counties in terms of having service centers. Descriptive-analytic method was applied to achieve the mentioned aim. Data was also collected using documentary and library methods. Data Envelopment Analysis (DEA) model is one of the effective ways to measure efficiency. The present study investigates the performance and efficiency of 17 counties in West Azerbaijan province in terms of having service centers using CCR and BCC models in data envelopment analysis method. The results indicate only 6 counties on CCR model and 7 counties on BCC model are relatively efficient and the remaining ones are inefficient. Finally, inefficient counties are modeled and the improvements needed for them to meet the efficiency boundary are explained.

K e y w o r d s: Service centers, efficiency evaluation, data envelopment analysis model, West Azerbaijan Province.

Introduction In recent decades, urbanization has been preferred on challenges for counties (Chen, 2014). One of the municipal engineering so urban growth is faced serious challenges faced by urban and regional planners is the problems and challenges (Ebrahimzadeh and Rafiee, uneven and unbalanced growth of counties and regions. 2009: 123). It is for the first time in history that more By looking at the spatial distribution of services in than a half of the world's population lives in urban counties and urban areas it can be argued that services areas. It is predicted that urban population will be have been distributed in an unbalanced manner duplicated in developing countries by 2030. This rapid (Taqvaii & Akbari, 2009: 97). Urban development can population development brings difficulties and major be sustained if it provides specific strategies and Efficiency measurements on service centers using data envelopment analysis model… 28 approaches to optimally meet residents’ service needs. Due to their attitudes, structural weaknesses in urban Theory and Methodology management and the lack of public participation, The present study is an applied one with mathematical service providers have not been able to efficiently modeling approach. The mathematical model used in distribute spatial balanced service. However, the focus this study is data envelopment analysis including BCC of service centers on a single space not only results in and CCR with an output-based approach. The variables high and low bipolar areas but also inflames the used in the study are extracted and defined according consumer population flow into these counties, which in to the data collected on 2014 of counties in West turn leads to environmental pressure, traffic, pollution Azerbaijan province. The data were analyzed using including sound and air etc. It is worth mentioning that DEA model in Frontier Analyst software and the this phenomenon attracts complementary and parallel results were extracted. Then, using efficiency scores, applications, and exacerbates the polarization of spatial counties were ranked. Finally, counties are modeled by counties (Khakpour & Bavanoori, 2009: 187). The introducing a model county to the inefficient counties. inequality and spatial imbalance in counties is not a Level of improvement in outputs needed for inefficient new phenomenon in the world, although in developing counties to reach the efficiency threshold is explained countries, due to the significant socioeconomic and as well. economic disparities and the imbalance in the distribution of services, the spatial differences of Research technique counties have been intensified and is more pronounced This method, which is globally known as efficiency (Abdi Daneshpour, 1999: 37). In Iran, like other measurement method provides a breakdown for outputs countries, the inappropriate distribution of urban while measuring the efficiency of return to the scale in services in different counties is highly concerned, and production. Currently, DEA is one of highly researched has become a transnational problem. The main concern areas in measuring performance which has been widely of urban authorities has been the provision of urban addressed by researchers around the world. This services and less attention has been paid to its proper method is applied to assess the performance of public distribution (Kamran et al., 2010: 148). The purpose of and nonprofit organizations whose pricing information this study is to scale counties in West Azerbaijan is usually unavailable or unreliable. This method is regarding providing service centers and identifying the generally uses the term Decision Maker Unit (DMU) level of their efficiency in services. Data envelopment instead of the term producer in order to be committed analysis model has been applied since DEA model is a to generalization. DEA uses a linear programming nonparametric approach. It also provides possibility to technique and is among nonparametric methods for rank populations and analyze the contribution of each estimating isoquant functions (Imami Meybodi, 2000). indicator (Charnes et al., 1977: 430). This model is Input- or output-based models: Given that the inputs of based on a linear programming approach and its main business organizations are centrally determined at a purpose is to compare and evaluate a number of certain level, the output-based model (output decision makers having different inputs and outputs maximization derived from the specified input) is used (Afshar Kazemi et al., 2006: 42). This study applies in this study. two basic models of data envelopment analysis including CCR model and BCC model. Counties of the CCR model study province are ranked according to the service This model, which is a planning pattern, seeks to centers indicators. maximize the relative efficiency of the unit p by

29 Journal of Geography and Spatial Justice. Year 1th - Vol. 1 - Issue. 2 - Spring 2018 choosing a set of weights for all inputs and outputs, equal to 1. while the rating of each unit should be less than or Equation 1 For DMUp functionality to be only determined using a linear programming method, program (2) has been developed as follows: Equation 3

Where wp is the relative efficiency of the decision-making unit (DMUp). In other words, it seeks to maximize output with respect to the institutional constraints. Xi and yr represent k input and s output for where, and *are dual variables. n units, respectively. V and u vectors also show the DMUp is efficient if weights of inputs and outputs, respectively. The first limitation is actually the denominator of the primary and . In order to objective function of the fraction, which allows the ensure that we do not have any weights of 1, and all model to be solved in a linear programming inputs and outputs can be achieved in the solution of framework. The second limitation ensures that, under the model, the program (1) is corrected using ε as the selected set of weights, the efficiency score of any follows, which is usually considered as small as 0.001 one of the decision units is not greater than 1. The or 0.0001: above model must be implemented for each decision Equation 4 unit to determine the relative efficiency of each unit. According to the literature, it was found that in the CCR model, if the number of units does not differ significantly from the inputs, most units will usually be effective or be placed on the efficiency boundary. A dual model is used to solve this problem. Equation 2 In this model, both base input and output are assumed to produce a constant return, i.e. if the inputs are doubled, outputs are doubled too (Martinez & Waldo, 2014: 5)

BCC model:

DMUp is efficient if and only if (1) and (2) This model is achieved by adding convexity constraints are required in model (2) (surplus variables constraint to CCR initial linear are equal to zero) and (Cooper et al,2004:12). programming so that the return to scale can be

constant, increasing, or decreasing. This constraint in

Efficiency measurements on service centers using data envelopment analysis model… 30

CCR model will cause the new variable (u) to appear in (Michailova et al 1996). Sampaio and Stosic, using BBC model. DEA method, estimated technical efficiency of 4,796 Equation 5 municipalities in Brazil (Sampaio & Stosic, 2003). Gonza'elz et al. (2011), in a study entitled “The importance of geographic analysis in life quality assessment: A case study of Spain”, measured the efficiency of life quality indicators at three geographical levels (region, province and county) using Data Envelopment Analysis Method. In their study of "Measuring life quality in Canary islands"

Martin and Mendoza (2013), using the DEA model, Equation 5 measured 19 life quality indicators in 87 counties of the Canary Islands in Spain. Poldaru and Roots (2014), in "The DEA model, a method for estimating life quality in Estonia," measured life quality in Estonian counties and analyzed 15 counties regarding their statistical indicators during 2000-2011. They rated

counties using DEA model. The BCC model can also be based on outputs. In Iran, Akbari and Basiri Parsa (2003) measured technical efficiency of municipality It is just needed to add constraint to the development activities in urban areas using DEA initial planning of CCR baseline output (Keshavarz & method (Akbari et al., 2003). Zaryari et al. (2010) Toloo, 2014: 452). evaluated the developmental efficiency of 30 provinces

in Iran using DEA. Based on their findings, ten Literature review provinces were efficient. However, the present study is Data envelopment analysis was first introduced in the the first attempt to address performance efficiency of doctoral thesis of Charens guided by Cooper and counties in West Azerbaijan Province regarding Rhodes. It was first used to evaluate the relative distribution of service centers. efficiency of US National Schools and published in

1978. This model was named after its providers to The study area CCR model (Charnes, 1978: 433). In 1984, Bunker, West Azerbaijan Province, including Lake, Charens and Cooper published a paper in which a covers an area of 43,660 square kilometers. The model called BCC was introduced. The new model province, which is located in the northwest of Iran, included return to scale (Vadodi Mofid, 2005). The accounts for 2.65% of the total area of the country and literature shows that this method has been applied in is located between 35 degrees 58 minutes to 39 degrees several areas, but there is no report on measuring the and 46 minutes northern latitude and 44 degrees 3 efficiency of service centers. Worthington et al. minutes to 47 degrees 23 minutes East latitude. The measured the efficiency of 103 Australian local province is adjacent to Azerbaijan and on the governments in internal management and recycling north and northeast, and to the west, services using DEA method (Worthington & Dollery province to the south and eastern Azerbaijan 2001). Michailove et al. measured the efficiency of 24 and Zanjan provinces to the east. According to the Bulgarian municipalities using DEA method latest statistics and divisions, the province has 17

31 Journal of Geography and Spatial Justice. Year 1th - Vol. 1 - Issue. 2 - Spring 2018 counties, 36 districts, 109 rural districts and 3728 Center, Iran, 2015). villages. Its population is about 326, 5219 (Statistics

Figure 1: Geographic location of under studied area (Source: Authors, 2018)

Results and Discussion Urban transport services (including transportation Public services refer to those services that provide services, etc.), (Study Center of Urban and Rural residents with needed facilities, and facilitate the Services, 2006: 3). production process which do not produce tangible The decision-making units in data envelopment goods. According to the definition of the Iranian analysis are considered to be systems that convert Statistical Center and the facts of urban life, services inputs into outputs. A system that is able to produce are classified as follows: more outputs using fewer inputs is more efficient, and Social services (including health services, educational has better performance. By providing a systematic services, etc.). approach to decision-making units, data envelopment Welfare and leisure services (including artistic, analysis models aim to maximize their efficiency (Ziari recreational, sports services). et al., 2010: 262). Reception and accommodation services (including hotel, restaurant, etc.) Higher Education Information services (including computer, print, visual The socio-economic development of a country is and audio services, etc.) significantly influenced by literacy indicator. In 2011, Financial and commercial services (including banking, West Azerbaijan province had 75 higher education insurance, etc.) institutions. Urmia, and Miyandoab are the most

Efficiency measurements on service centers using data envelopment analysis model… 32 populated counties in west Azerbaijan, respectively. As With a quick look at the spatial distribution of various in Table 1, 25.3% of the total number of institutions is health services in the West Azerbaijan province, Urmia located in Urmia. Khoy with 10.6%, Miyandoab with has the biggest number of hospitals, hospital beds and 10.6 and with 6.6% are followed. Followed doctors working in the province which followed by by Urmia, Khoy and Miyandoab have a higher Khoy and Miyandoab. Three newly-established percentage of higher educational institutions than other counties of Chaypare, Poldasht and Shoot are in the counties in the province. Therefore, it can generally be lowest ranks, respectively. Cold storage rooms, silo stated that at the provincial level, higher educational and warehouses distribution. In 2014, West Azarbaijan institutions are concentrated in the provincial capital. province has 15 public warehouses, 4 silos and 122 Health centers (Hospitals) based on number of beds cold storages. Urmia has the largest number of and doctors warehouses.

Table 1: Distribution of services (higher education, medical centers, cold storages warehouses and silos in the counties County Cold storages, Health Centers Higher Education Institutes Warehouses and Silos

Cold Storage Cold Vocational& Vocational&

Hospital No Hospital

Doctors No

Warehouse Vocational Vocational Non-Profit other UNI

State UNI PNU UNI Institutes zdUIAzad UNI Bed No Bed

UNI Silo

85 3 2 760 1786 11 4 1 2 4 6 2 Urmia Oshnaviyeh 3 0 0 20 25 1 0 0 1 0 0 0 Bookan 2 1 0 65 133 1 0 1 1 0 0 2 Poldasht 0 0 0 13 20 1 0 0 1 0 0 0 Piranshahr 0 0 0 21 62 1 0 1 1 0 0 2 Tekab 0 0 0 23 73 1 0 1 1 0 1 1 Chaldoran 0 0 0 19 25 1 0 1 1 0 0 0 Chaypare 0 0 0 - 30 1 0 1 1 0 0 0 Khoy 4 1 1 79 342 2 0 2 2 2 0 2 Sardasht 0 0 0 34 117 1 0 1 1 0 0 1 4 2 0 80 165 1 0 1 2 0 0 1 Shahindezh 1 0 0 33 87 1 0 1 1 0 0 1 Shoot 0 0 0 17 10 1 0 0 1 0 0 0 Mako 1 1 0 35 120 2 0 1 1 0 0 1 Mahabad 8 4 1 111 244 1 0 1 1 0 1 2 Miyandoab 5 1 0 112 252 2 1 1 2 0 2 2 Naghade 9 2 0 70 142 2 0 1 1 0 0 1 Total 122 15 4 1492 3632 31 5 15 21 6 10 18

Hotel and hospitality centers distribution In spite of many cultural, historical and, in particular centers are concentrated in Urmia which is followed by natural, cultural heritages in the region, the border with Khoy and Mako. Chaypare, Shoot and Chaldoran have three foreign countries has doubled the touristic the lowest rank in hotels and hospitalities. attractions in this region. The importance of tourism in the economic dimension is such that other economic activities are organized in accordance with the demands of the tourism market. Cultural service distribution Considering the statistics obtained in 2015 Cultural service centers are poorly and inefficiently (Table 2, the largest number of hotels and hospitality distributed. Based on statistics and data obtained in

33 Journal of Geography and Spatial Justice. Year 1th - Vol. 1 - Issue. 2 - Spring 2018

2015, the majority of these centers are located in have the highest number of cultural centers in the Urmia. The following tables show the number of province. Poldasht, Shoot and Chaypare are in the centers providing cultural services in the West lowest ranks, respectively. Azerbaijan province. Urmia, Khoy and Miyandoab

Table 2: Distribution of the services (tourist centers and cultural services centers) in the counties Cultural Services Centers Tourist centers

County Press

Offset THeaters Digitala Bindery

The Movies Movies The

News stand stand News Hotels & Inns

Lithography

TAMPO Print

Advertising Agencies Advertising HeligraverFlexo & Restaurants & Dinners Dinners & Restaurants Cultural products stores

Certificated Publications

6 6 10 6 34 32 129 3 135 94 38 2 1168 87 Urmia Oshnaviye 0 0 1 0 1 1 11 1 4 3 0 1 59 2 Bookan 0 1 0 1 4 1 33 1 6 5 3 2 183 6 Poldasht 0 0 0 0 1 0 0 0 1 0 0 0 32 0 Piranshahr 0 0 0 0 1 1 24 0 3 8 0 1 89 3 Tekab 0 0 0 0 1 1 13 0 1 7 1 1 54 1 Chaldoran 0 0 0 0 0 1 10 0 1 3 1 1 24 0 Chaypare 0 0 0 0 1 1 0 0 1 1 1 1 13 0 Khoy 0 1 3 2 3 13 32 3 15 10 9 1 717 24 Sardasht 0 0 1 0 1 1 13 1 4 1 0 1 130 6 Salmas 0 0 6 1 3 4 26 1 5 6 2 1 287 6 Shahindezh 0 0 0 1 1 1 37 1 2 3 1 1 79 3 Shoot 0 0 0 0 1 0 0 0 1 2 0 0 22 0 Mako 0 0 1 0 3 4 30 1 4 10 2 1 290 17 Mahabad 0 1 6 0 3 5 19 1 10 8 2 1 174 11 Miyandoab 0 0 1 0 4 7 84 1 7 7 3 1 268 7 Naghade 0 0 4 1 2 3 23 1 4 11 3 2 155 5 Total 6 9 33 12 64 77 480 15 204 179 64 18 3744 178

Banking unit’s distribution The majority of state-owned commercial banks are Transportation services distribution located in , especially in Urmia County. Based on data and statistics, the majority of passengers In addition, the diversity of banks is higher in Urmia. and transportation services (passengers, transportation Urmia, Khoy and Miyandoab have the greatest number companies, freight forwarding companies, public of banks, respectively. Poldasht, Chayipare and Shoot terminals and public cargo terminals) are located in are at lowest ranks, respectively. Urmia which is followed by Mako and Khoy. Shoot.

Efficiency measurements on service centers using data envelopment analysis model… 34

Chaldoran and Poldasht are at lowest ranks, Urmia, followed by Khoy and Miyandoab. Shoot, respectively. Chaypare and Poldasht are in the lowest ranks, Postal services distribution respectively. According to the analysis, the highest number of postal service providers with 31.6% is located in

Table 3: Distribution of the services (banking and transportation services) in the counties County Transportation services centers Banking service centers

Tose Saderat public cargo

Keshavarzi forwardin passengers Post Bank

terminals terminals

Maskan Saderat

Tejarat freight public Sepah Refah Melat Sanat Meli g

Urmia 2 1 63 30 4 1 25 1 17 14 15 29 37 36 38 Oshnaviye 0 1 3 2 0 0 1 0 1 1 1 1 1 1 1 Bookan 0 1 8 5 1 0 4 0 3 3 2 3 5 3 3 Poldasht 0 1 2 1 0 0 1 0 0 0 0 0 0 1 0 Piranshahr 0 1 3 5 0 0 1 0 1 1 1 1 1 1 2 Tekab 0 1 2 5 1 0 1 0 1 1 1 1 1 1 2 Chaldoran 0 1 0 1 0 0 1 0 1 2 1 1 1 1 1 Chaypare 0 0 4 2 0 0 1 0 1 0 0 1 0 1 0 Khoy 0 1 12 11 1 0 5 1 4 10 5 9 7 9 13 Sardasht 0 1 2 3 1 0 1 0 1 1 1 1 2 2 2 Salmas 0 1 6 11 2 0 3 0 2 7 2 4 2 2 6 Shahindezh 0 0 3 2 1 0 1 0 1 4 1 2 1 2 3 Shoot 0 1 0 1 0 0 1 0 1 0 0 1 0 1 0 Mako 0 1 18 5 0 0 3 0 1 6 3 2 4 2 8 Mahabad 0 1 9 10 1 0 2 0 4 3 4 3 4 2 4 Miyandoab 0 1 10 7 1 0 2 0 3 4 4 4 3 7 5 Naghade 0 1 5 8 1 0 1 0 3 3 2 3 2 2 4 Total 2 15 150 109 14 1 54 2 45 60 43 66 71 74 92

Efficiency measurement based on CCR and BCC models CCR and BCC models are two basic models in data This study utilizes both models to measure the envelopment analysis. CCR measures efficiency by efficiency of service centers in counties of West assuming a constant return to scale and BCC measures Azerbaijan province. efficiency on the assumption of variable returns to scale. These models have two applications, one is based on minimizing the use of production factors and the other is based on maximizing production factors.

35 Journal of Geography and Spatial Justice. Year 1th - Vol. 1 - Issue. 2 - Spring 2018

Figure 2: Ranking of counties based on BCC model

Figure 3: Ranking of counties based on the CCR model

As seen in the figures above, 6 counties in counties, while Naghadeh can only be modeled on 2 CCR and 7 counties in BCC are efficient. Figures 4 counties. It is argued that (according to Figure 4) and 5 show effective areas based on CCR and BCC. inefficient counties are needed to follow the model of The number of inefficient counties for which an efficient counties which is 6 according to the analysis efficient county can serve as a model to achieve performed by CCR which is 6 according to the analysis efficiency threshold has been identified. Figure 4, performed by CCR. based on CCR model, shows that Khoy and Urmia are the most efficient counties and can be modeled for 10

Efficiency measurements on service centers using data envelopment analysis model… 36

Figure 4: Efficiency model based on CCR model

In the following figure, efficient counties are counties such as Naghadeh and Salmas serve as a identified based on BCC model. 41% of counties are model for only 4 counties. Accordingly, based on the efficient. The results of this model show Urmia can act analysis, it is better to apply these seven counties as as a model for 11 inefficient counties, while other model counties in BCC.

Figure 5: Efficiency model based on BCC model

Conclusion identify efficient and inefficient counties in order to Establishing service centers in counties is among the improve the status quo, to move towards the efficiency, main elements in improving social welfare. Nowadays, to improve service centers in the province and to the majority of counties not only face a shortage in distribute them in a balanced manner. Therefore, using service centers but also an unbalanced distribution of DEA model, through proper modeling from other services. In turn, efficiency measurement of counties efficient counties, with the allocation of appropriate regarding establishing service centers is of great resources and policies, it is possible to promote other importance. Hence, it is essential to improve the counties and bring them to the efficiency frontier. efficiency of urban services, while distributing services in a way by which all counties are provided with References balanced service. This study measures the efficiency of Ebrahimzadeh, I. & Rafiee, Q. (2009). An analysis counties in West Azerbaijan province regarding on the spatial distribution pattern of Marvdasht municipal services using CCR and BCC models. using Shannon and Holdern entropy models and According to CCR model, 35.29% and according to presenting its future expansion pattern, Human BCC model, 17.14% of counties are efficient, and the Geography Research, 69, 138-123. remaining are inefficient and do not utilize their Kazemi Afshar, M. A., Setayesh, M. R., available resources and facilities in an optimal manner. Mehrabian, S. & Aliyanori, K. (2006). Evaluation Efficiency does not concern the amount of resources of the relative efficiency of branches of Saderat available, but also how to properly utilize and manage Bank, Iran using Data Envelopment Analysis the facilities. The purpose of ranking counties is to

37 Journal of Geography and Spatial Justice. Year 1th - Vol. 1 - Issue. 2 - Spring 2018

Model (DEA), Bank and Economics Journal, 57, Charnes, A., Cooper, W. W. & Rhodes, E. (1978). 48-42. Measuring the efficiency of decision-making units, Akbari, N. & Nayereh Basiri, P. (2003). European Journal of Operational Research, 2, 429- Measuring the Technical Efficiency of Urban 444. Development Activities Using DEA Method in Chen, M. (2014), Sustainable urban land use Hamedan Province, Sustainable Growth and planning, WBI Urban Practice. Development Researches, 17, 154-133. Gonzalez. E., Carcaba, A. & Ventura, J. (2011). Emami Meybodi, A. (2000). Principles of The Importance of the Geographic Level of Measuring Productivity and Efficiency, , Analysis in the Assessment of the Quality of Life: Institute of Business Studies, First Edition. The Case of Spain, Soc Indic Res, 102, 209–228. Taqvaii, M. & Akbari, M. (2009). Spatial Analysis Keshavarz, E. & Toloo, M. (2014). Finding of Development Indicators in Iranian Regional efficient assignments: An innovative DEA Municipalities, Geographical Research Journal approach, Measurement, 58, 448–458. (Iranian Geographical Community), 20, 111-97. Martinez, M. & Nunez, W. (2014). Efficiency Khakpour, B. (2009). Analysis of inequality in the Analaysis of information Technology and Online levels of development in the regions of , social Networks Management: An integrated Journal of Knowledge and Development, 16(27), DEA- Model Assessment, Information & 202 182. Management, 51(6), 712–725 Ziyari, K., Sorkh Kamal, K. & Zanjirchi, M. Michailove, A. M. & Nenkova Tomova, P. (1996). (2010). Analysis of Classical Performance of the Cost Efficiency in Bulgarian Municipalities, Provinces in terms of Development using DEA University for National and World Economy, Method, Planning and Approximation of Space, Sofia. 14(3), 257-265. Põldaru, A., Vitae, R. & Jüri, R. (2014). A PCA– Abdi Daneshpour, Z. (1999). Analysis on the DEA approach to measure the quality of life in Spatial Imbalance in the counties, Tehran, Tehran, Estonian counties, Socio-Economic Planning Safeh Journal, Ninth, No.29, Shahid Beheshti Sciences, 48(1), 65–73. University of Tehran. Sampaio de Soua. M. C. & Stosic, B. (2003). Kamran, H., Parizadi, T. & Hosseini Amini, H. Technical Efficiency of the Brazilian (2010). Leveling of Urban Services in Municipalities: Correction Non-Parametric Metropolitan Regions in Tehran, Journals of Frontier Measurements for Outliers, Department of Geography and Regional Planning, 1(1), 147-164. Economic. University of Brazil. Journal of Vadodii B., Mosa Khani, M. & Hamidi N. (2006). Productivity Analysis, 24(2), 157-181. Developing a Methodology for Assessing Cooper. W., Lawrence, W., Seiford, M. & Joe, Z. Efficiency and Productivity in the Higher (2004). Handbook on Data Envelopment Analysis, Education Section, Journal of Researcher International Series in Operations Research (Management), 3(6), 98-121. Management Science, 164. Mendoza, C. & Juan Carlos, M. (2013). A DEA Worthington, A. & Brian, D. (2001). Measuring Approach to Measure the Quality-of-Life in the Efficiency in Local Government: An Analysis of Municipalities of the Canary Islands, Social New South Wales Municipalities, Policy Studies Indicators Research, 113, 335-353. Journal, 29(2), 232-249.